TabPy (the Tableau Python Server) is an Analytics Extension implementation which expands Tableau's capabilities by allowing users to execute Python scripts and saved functions via Tableau's table calculations.

Consider reading TabPy documentation in the following order:

• About TabPy
• TabPy Installation Instructions
• TabPy Server Configuration Instructions
• Running TabPy in Virtual Environment
• Running TabPy on Heroku
• Authoring Python calculations in Tableau.
• TabPy Tools

Important security note:

• By default, TabPy is configured without username/password authentication. We strongly advise using TabPy only with authentication enabled. For more information, see TabPy Server Configuration Instructions. Without authentication in place, if the TABPY_EVALUATE_ENABLE feature is enabled (as it is by default), there is the possibility that unauthenticated individuals could remotely execute code on the machine running TabPy. Leaving these two settings in their default states together is highly discouraged.

Troubleshooting:

• TabPy Wiki

More technical topics:

• Contributing Guide for TabPy developers
• TabPy REST API
• TabPy Security Considerations

Other useful resources:

• Tableau Sci-Fi Blog provides tips, tricks, under the hood, useful resources, and technical details for how to extend Tableau with data science.
• Known Issues for the Tableau Analytics Extensions API.
• For all questions not related to the TabPy code (installation, deployment, connections, Python issues, etc.) and requests use the Analytics Extensions Forum on Tableau Community.
• Building advanced analytics applications with TabPy
• Building Data Science Applications with TabPy Video Tutorial
• TabPy Tutorial on TabWiki